Lithium Batteries are SOOO yesterday

[mrm]

Originally Posted by kas_1611

The efficiency is vastly improved when you factor in the calorifier in the system that uses the waste heat from the engine to give you the tank full of hot water for your shower and the alternator that converts some of that rotational energy back into electrical energy to charge your batteries. So in that respect the humble marine diesel is more energy efficient than it's land based cousin.

Keiron

Actually I had a remark on calorifier ready to include in my post, but realised it was not really applicable and deleted it not to cloud the issue. On our boat, calorifier will reach maximum temperature in less than an hour of engine run. After that time no further water heating in calorifier takes place and we are back to original efficiency of a Diesel engine.

BTW, same principle is used on land, on large scale. Many electric power plants are located close to large urban areas to provide houses and industry with heat as well as electricity.

As for alternator you are wrong, sorry. Alternator consumes mechanical energy produced by an engine, not a waste heat, so I do not see how it could improve thermodynamic efficiency of a Diesel engine.

[K_V_B]

Originally Posted by mrm

Actually I had a remark on calorifier ready to include in my post, but realised it was not really applicable and deleted it not to cloud the issue. On our boat, calorifier will reach maximum temperature in less than an hour of engine run. After that time no further water heating in calorifier takes place and we are back to original efficiency of a Diesel engine.

BTW, same principle is used on land, on large scale. Many electric power plants are located close to large urban areas to provide houses and industry with heat as well as electricity.

As for alternator you are wrong, sorry. Alternator consumes mechanical energy produced by an engine, not a waste heat, so I do not see how it could improve thermodynamic efficiency of a Diesel engine.

And actually a calorifier reduces the thermodynamic efficiency of your engine... "Combined Heat and Power" is actually not at all that efficient... The most efficient way to turn "chemical energy" in to heat is actually to turn it in to electricity first, and then power a heat pump with it. That would be to complex for a boat however.

[Cadence]

Originally Posted by mrm

Actually I had a remark on calorifier ready to include in my post, but realised it was not really applicable and deleted it not to cloud the issue. On our boat, calorifier will reach maximum temperature in less than an hour of engine run. After that time no further water heating in calorifier takes place and we are back to original efficiency of a Diesel engine.

BTW, same principle is used on land, on large scale. Many electric power plants are located close to large urban areas to provide houses and industry with heat as well as electricity.

As for alternator you are wrong, sorry. Alternator consumes mechanical energy produced by an engine, not a waste heat, so I do not see how it could improve thermodynamic efficiency of a Diesel engine.

I believe the power plants providing heat, mainly to businesses or high rises, are steam plants.

[SV THIRD DAY]

Originally Posted by Cadence

I believe the power plants providing heat, mainly to businesses or high rises, are steam plants.

True...but they are typically co-generation plants.
First they use the steam to turn a steam turbine to generate electricity and then use the "waste heat" to send to the businesses. It would be rare to just generate steam as a stand alone for heating, it's usually a "waste product" of power generation.

[Cadence]

Originally Posted by SV THIRD DAY

True...but they are typically co-generation plants.
First they use the steam to turn a steam turbine to generate electricity and then use the "waste heat" to send to the businesses. It would be rare to just generate steam as a stand alone for heating, it's usually a "waste product" of power generation.

I can see how you might think that I meant steam was the primary product not electricity. Poor phrasing on my part.

The only reason I brought it up was I didn't see its relevance to diesel and boats?

[mrm]

Originally Posted by Cadence

The only reason I brought it up was I didn't see its relevance to diesel and boats?

This post http://www.cruisersforum.com/forums/...ml#post1777175 by kas_1611 should help.

Originally Posted by K_V_B

And actually a calorifier reduces the thermodynamic efficiency of your engine...

Why? Calorifier is usually connected to the 'big loop' in engine coolant flow, so it does not influence engine's operating temperature and utilises only waste heat. How come it reduces the thermodynamic efficiency of an engine then?

[Cadence]

Originally Posted by mrm

This post http://www.cruisersforum.com/forums/...ml#post1777175 by kas_1611 should help.



Why? Calorifier is usually connected to the 'big loop' in engine coolant flow, so it does not influence engine's operating temperature and utilises only waste heat. How come it reduces the thermodynamic efficiency of an engine then?

Hell, put a heat exchange in the raw water discharge. The heat is only going overboard at that point.

[mrm]

Originally Posted by Cadence

Hell, put a heat exchange in the raw water discharge. The heat is only going overboard at that point.

Yep, that's why I am curious...

[Adelie]

Originally Posted by zboss

Also, I think most of the major crashes (air france, the latest Malaysian air crash, etc) have all been attributed to human error. The computers knew what to do but were not allowed to do it. If the pilots had just let the plane fly itself then this would not have happened. Why was Air France not on autopilot anyhow?

Originally Posted by Adelie

The autopilot disengaged because it had experienced conditions beyond its programming.

My original point was that the plane would have also crashed even if the pilots let the plane continue to fly itself.

Originally Posted by zboss

The autopilot did what it was supposed to do.

The autopilot was supposed to fly the plane. It didn't do that, it quit. There could be a long contentious semantic argument to rival the RightOfWay/StandOn war currently underway on the COLREGS thread. I'll just say that I can see your view on this up to a point, can you see mine at least a little bit here?

Originally Posted by zboss

There is no such thing as being "outside its programming". The autopilot programing said "if x then Y" and it did it. The air speed indicators got iced up. The fact that the pilot was the least experienced crew member says a lot about their training, not about the autopilot. What did the PILOT do when the air speed indicators iced up... he immediately did the wrong thing. At least the computer got it right.

The more autopilots are put in charge of planes, the smarter they get. There are only so many obstacles that exist in flight and I do not believe there is a discipline that humans can do that computers can not do better.

Autopilots aren't getting smarter, they're getting more complex programming that can deal with less common situations.. More complex also means easier to break and more subject to bugs and unforeseen interactions between programming modules. Also the programming can only address the situations that the programmers can think of, if something happens outside of the programming the can't think critically about all the symptoms to determine what the problem is.

I will agree that the proximate cause of the AF447 crash was human error, the pilot should have sought confirmation that the plane was overspeeding, rather than holding to that assumption all the way into the ground. So the airspeed indicator is non-functioning, I expect that the GPS would give a speed over the ground. While that won't you airspeed, and won't let you determine if you are stalling or overspeeding at altitude, as the plane goes lower the mach limit and stall speed diverge and the GPS should let you figure out which is your problem.

That said there are a number of underlying contributing causes:
1) Deterioration of flight skills in the era of automation which can be laid on the airlines and the regulatory agencies.
2) Lack of tactile feedback in the control stick and the design decision to average the inputs. This prevented the other pilots from figuring out that the goofball was not doing as he was told: to lower the aircraft nose. This can be laid on Airbus.
3) The FAA is currently changing testing requirements for stall recovery because they are realizing that it was contributing to bad habits on the part of pilots. Being French these pilots were not subject to the FAA, but I suspect the EASA and DGCA may have similar issues.

The days of pilots are not numbered, pilotless airliners are decades away. In the meantime airlines and regulatory agencies need to find more effective ways to maintain pilot skills.

[goboatingnow]

That said there are a number of underlying contributing causes:
1) Deterioration of flight skills in the era of automation which can be laid on the airlines and the regulatory agencies.
2) Lack of tactile feedback in the control stick and the design decision to average the inputs. This prevented the other pilots from figuring out that the goofball was not doing as he was told: to lower the aircraft nose. This can be laid on Airbus.
3) The FAA is currently changing testing requirements for stall recovery because they are realizing that it was contributing to bad habits on the part of pilots. Being French these pilots were not subject to the FAA, but I suspect the EASA and DGCA may have similar issues.

The days of pilots are not numbered, pilotless airliners are decades away. In the meantime airlines and regulatory agencies need to find more effective ways to maintain pilot skills.

I read all three reports

There was no indication in any of the three comprehensive reports that sidesticks were in any way a contribution to the issue. modern centre and side sticks all essentially have computerised feedback these days anyway.

The primary issue with AF447 was two fold

(a) lack of warning or display of excessive angle of attack. The instruments simply " refused" to display parameters that were so far out of range

(b) Lack of training on high speed high altitude stalls, note the aircraft was virtually unrecoverable after that . Air france has subsequently instituted such training I believe

The primary cause was the failure of the airspeed pitots, combined with the selective display of information leading to the pilots not recognising the situation they were in.

What AF447 shows is the very complex issues surrounding man machines interfaces in an era of automation. We are likely to see such issues actually increase as we employ more normal automation, but rely on humans to get us out of trouble. We're a very very long way from allowing the computers to " get us out of trouble"

Originally Posted by zboss
There is no such thing as being "outside its programming". The autopilot programing said "if x then Y" and it did it. The air speed indicators got iced up. The fact that the pilot was the least experienced crew member says a lot about their training, not about the autopilot. What did the PILOT do when the air speed indicators iced up... he immediately did the wrong thing. At least the computer got it right.

Thats not true, the fact is the programming of the autopilot could not cope with the lack of speed inputs and in effect returned control to the pilot ( Im simplifying ), hence the programmers did not build in enough resilience , this isn't uncommon

Were a long long way from duplicating the problem solving ability of the human brain

dave

[zboss]

Originally Posted by goboatingnow

Thats not true, the fact is the programming of the autopilot could not cope with the lack of speed inputs and in effect returned control to the pilot ( Im simplifying ), hence the programmers did not build in enough resilience , this isn't uncommon

I guess we can agree to disagree here rather than fight. The system recognized an issue that was not programmed for and correctly disconnected rather than making a fatal error, this is not even a system crash, which is also a programmed solution. The autopilot did not crash the plane, a human did. It said... "hm... something is wrong here I better inform the pilots and disconnect."

The speed indicator would have been on the plane regardless of the presence of an autopilot or not. The indicator would have frozen either way. In a way, a bad design of the speed indicator is at fault along with a lack of redundant speed indicators. I assume a very large plane must have a more than one speed indicator.

[goboatingnow]

Originally Posted by zboss

I guess we can agree to disagree here rather than fight. The system recognized an issue that was not programmed for and correctly disconnected rather than making a fatal error, this is not even a system crash, which is also a programmed solution. The autopilot did not crash the plane, a human did. It said... "hm... something is wrong here I better inform the pilots and disconnect."

The speed indicator would have been on the plane regardless of the presence of an autopilot or not. The indicator would have frozen either way. In a way, a bad design of the speed indicator is at fault along with a lack of redundant speed indicators. I assume a very large plane must have a more than one speed indicator.

Like many accidents, there is no , one overriding reason, there was a typical " fault cascade ", resulting in " inexperienced pilots" being misled into believing something was , what it was not.

The fault was not redundant speed indicators, the plane had three, however the " voting " system acted on its logic to display what it display and to cause the autopilot to react the way it did.

You could take two simplistic conclusions

(a) The automation was at fault, as it did not cope with the fault. it had alternative sources of data and didn't use them , but turned control over to the humans to decide

(b) The humans were at fault , in that the misunderstood the information being presented to them and in their inexperience of such situations assumed a circumstance that was in effect not occurring


The correct answer is that the problem was a complex interaction or man/machine issues and human weakness. Until we have machines that can handle every reasonable conceivable solution , humans will be the final arbiter and as we know , we f^&k up every now and then.

The main issue in designing machine systems that can cope, is the tremendous problem, that software is written by humans and therefore suffers from humans inability to see all possible issues ( or to then build those into complex systems ). The great thing about humans over current or near future AI, is that while humans cannot foresee all the issues, they can take rapid action when supplied with partial information and new circumstances, without necessary having a worked out logical plan beforehand ( which computers need)

This fact , ( fizzy logic so to speak) continues to dog AI.

[Saltyhog]

Fizzy Logic... The logic I follow when there is not enough "fizz" in my Gin and Tonic. Later on I begin to feel "fuzzy", yet another logical abstraction.

[Cadence]

An aircraft thread?? Although some of what I read is about autosailing? I want everything to feed to my lap top, so I can sail? I guess if it lights your fire. What happens when it takes a dump? Can you get home?

[Adelie]

Originally Posted by zboss

I guess we can agree to disagree here rather than fight. The system recognized an issue that was not programmed for and correctly disconnected rather than making a fatal error, this is not even a system crash, which is also a programmed solution.

This little digression started out with 2 comments:

Originally Posted by zboss

I'm not a pilot but I feel that commercial pilots days are numbered. The computers fly the plane anyhow, . . .

And

Originally Posted by zboss

If the pilots had just let the plane fly itself then this would not have happened.

And we have further digressed into the semantics of whether the autopilot "failed" when it was no longer able to fly the plane (the intended purpose of the autopilot) or whether it successfully followed its programming (the intended purpose of the computer underlying the autopilot) and turned the aircraft back over to a human when it got out of its depth and realized it.

It doesn't really matter what the answer is, automation is nowhere near replacing pilots. Not only that automation is degrading the skills of the pilots that will take over when it fails/declines to fly the plane. So if airline pilots who are paid lots of money to maintain their skills and well heeled airlines who have huge liability incentives to have pilots maintain their skills have problems with this then what does that say about sailing where automation has lowered the bar to entry on top of everything else?



Adelie